Ride-Sharing Passenger Group Evaluation for Dynamic Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ride-sharing technologies fail to accurately determine optimal passenger combinations for on-demand ride-sharing scenarios, where users frequently change during travel, considering user attributes and maintaining preferences.
Innovation Solution
An information processing apparatus calculates a second evaluation value for user groups based on a first evaluation value representing user attributes and the duration of their travel together, determining response policies for candidate users to ensure preferences are met, such as gender balance or similarity, and sends service commands to autonomous vehicles to adjust pick-ups and drop-offs accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ride-sharing technologies are used, then the system can operate with basic functionality, but the accuracy of determining optimal passenger combinations is insufficient
Solution Approach 1:
The evaluation system is segmented into multiple independent evaluation values: first evaluation value based on user attributes, second evaluation value based on travel time together, and third evaluation value for overall optimization. This segmentation allows each aspect to be evaluated separately and then combined, improving accuracy without overwhelming complexity.
Solution Approach 2:
The system changes parameters by introducing multiple evaluation dimensions (attributes, travel time, compatibility) and dynamically adjusting their weights. The control unit calculates different evaluation values based on varying parameters such as user attributes and duration of travel together, enabling precise matching while managing complexity through parameterized evaluation.
2Measurement precision
If user attributes and travel duration are considered for preference calculation, then passenger matching accuracy improves, but the computational complexity increases
Solution Approach 1:
The preference calculation is segmented into distinct evaluation components: first evaluation value from user attributes, second evaluation value from travel duration, and third evaluation value as the综合 result. This segmentation allows the control unit to process each component separately using straightforward calculations, then combine them, thereby improving precision while controlling computational complexity.
Solution Approach 2:
The system performs preliminary evaluation by calculating first and second evaluation values before determining the final third evaluation value. This preliminary action breaks down the complex preference calculation into preparatory steps that are computationally simpler, allowing accurate preference assessment without overwhelming computational burden.
3Productivity
If the system dynamically adjusts passenger combinations based on evaluation values, then ride-sharing optimization improves, but the response time for determining policies increases
Solution Approach 1:
The dynamic adjustment process is segmented into sequential evaluation steps: calculating first evaluation value from attributes, then second evaluation value from travel duration, and finally third evaluation value for policy determination. This segmentation allows parallel or cached computation of individual components, reducing overall response time while maintaining optimization efficiency.
Solution Approach 2:
The system performs preliminary calculations of user attributes and travel duration evaluations before final policy determination. By pre-calculating first and second evaluation values and caching them, the system reduces the time required for dynamic adjustments, enabling efficient ride-sharing optimization without excessive response delays.
Data Source
AI summary
An information processing apparatus manages a ride-sharing system in which a plurality of users travels together on the same moving vehicle. The apparatus includes a control unit configured to calculate a second evaluation value for a user group traveling on the same moving vehicle, based on a first evaluation value based on a user's attribute and a time during which the user group is maintained, and in a case where there is a candidate user who is a candidate of a user newly riding in a predetermined moving vehicle, calculate the second evaluation value when a predetermined user group is formed for the predetermined moving vehicle, and determine a response policy for the candidate user based on whether or not the second evaluation value satisfies a predetermined policy.